This is, I believe, the edit distance, and therefore dependent on the scoring scheme used by the aligner. Is there some way to generally back out the number of mismatches from this? Also, bwa appears to include only mismatches, but bowtie includes insertions and deletions in its NM.
Getting number of mismatches from bam record
I'm trying to estimate the sequencing error rates (mismatches and indels) from a bam file.
I can get the number and length of insertions and deletions from the cigar string by counting the number and length of the "I" and "D" values.
How do I calculate the number of mismatches without going to the reference fasta file? We can assume the MD tag is present, but I haven't figured out how to actually parse it properly.
(I'm using python with pysam, if someone has example code somewhere.)
1 answer
Simply use the NM tag.
• 0 views
•
link
updated
by
Ram
•
written
by
Zev.Kronenberg
•
written
by
Zev.Kronenberg
Log in to answer this question.
More posts like this
-
Best alignment algorithm for mapping NGS reads containing deletions of highly variable size?
written by Andrew P •
-
filter bam file by summarizing tags
written by Hadar •
-
Alignment Score (AS) tag in minimap2
written by tim.booth •
-
why does qualimap2 shows different results for number of mismatches for bowtie2 and BWAMEM?
written by maria2019
-
In Cigar String, What Is The Difference Between 'M' And 'X'
written by debanjan.roy12 •
-
How to convert from a cigar string to extended cigar string?
written by omariqbal2017 •
-
SAM MD tag for consecutive insertions/deletions
written by marongiu.luigi
-
Per base statistics on an RNAseq bam file
written by tassa.saldi •
-
Getting edit distance from sam tags MD & NM: conceptual misunderstanding?
written by WouterDeCoster
-
Get Number Of Mismatches In Pysam Colourspace Aligment
written by User 8663 •
